DOI QR코드

DOI QR Code

A Naive Multiple Imputation Method for Ignorable Nonresponse

  • 발행 : 2004.08.01

초록

A common method of handling nonresponse in sample survey is to delete the cases, which may result in a substantial loss of cases. Thus in certain situation, it is of interest to create a complete set of sample values. In this case, a popular approach is to impute the missing values in the sample by the mean or the median of responders. The difficulty with this method which just replaces each missing value with a single imputed value is that inferences based on the completed dataset underestimate the precision of the inferential procedure. Various suggestions have been made to overcome the difficulty but they might not be appropriate for public-use files where the user has only limited information for about the reasons for nonresponse. In this note, a multiple imputation method is considered to create complete dataset which might be used for all possible inferential procedures without misleading or underestimating the precision.

키워드

참고문헌

  1. An Essay on the logical Foundations of Survey, Part One, in foundations of Statistical Inference Basu, D.;V. P. Godambe(ed.);D. A. Sproutt;(ed.)
  2. Sampling Technique (3rd ed.) Cochran, W. G.
  3. Proceedings of the Survey Research Methods Section of the American Statistical Association A new approach to imputation Cohen, M. P.
  4. Bayesian Methods for Finite Population Ghosh, M.;Meeden, G.
  5. Metrika v.41 A minimal complete class theorem for decision problems where the parameter space contains only finite many points Lee, S. -C.;Meeden, G. https://doi.org/10.1007/BF01895320
  6. Journal of American Statistical Association v.95 A decision theoretic approach to imputation in finite population sampling Meeden, G. https://doi.org/10.2307/2669401
  7. Bayesian Analysis in Statistics and Econometrics Essays in Horner of Arnold Zeller An approach to the problem of nonresponse in sample survey using Polyar posterior Meeden, G.;Bryan, M.
  8. Journal of American Statistical Association v.86 A noninformative Bayesian approach to interval estimation in finite population sampling Meede, G.;Vardeman, S. https://doi.org/10.2307/2290513
  9. Journal of American Statistical Association v.91 A semiparametric transformation approach to estimating usual intake distributions Nusser, S. M.;Carriquiry, A. L.;Dodd, K. W.;;Fuller, W. A. https://doi.org/10.2307/2291570
  10. Journal of American Statistical Association v.91 On variance estimation with imputed survey data Rao, J. N. K. https://doi.org/10.2307/2291637
  11. Biometrika v.57 Jackknife variance estimation with survey data under hot deck imputation Rao, J. N. K.;Shao, J. https://doi.org/10.1093/biomet/57.2.377
  12. Biometrika v.57 On finite-population sampling theory under certain linear regression models Royall,R.M. https://doi.org/10.1093/biomet/57.2.377
  13. Multiple Imputation for Nonresponse in Surveys Rubin, D. B.
  14. Journal of American Statistical Association v.81 Multiple imputation for interval estimation from simple random samples with ignorable nonresponse Rubin, D. B.;Schenker, N. https://doi.org/10.2307/2289225
  15. Analysis of Incomplete Multivariate Data Schafer, J. L.